Echocardiography-based AI detection of regional wall motion abnormalities and quantification of cardiac function in myocardial infarction
ObjectiveTo compare the performance of a newly developed deep learning (DL) framework for automatic detection of regional wall motion abnormalities (RWMAs) for patients presenting with the suspicion of myocardial infarction from echocardiograms obtained with portable bedside equipment versus standar...
| Main Authors: | , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2022-08-01
|
| Series: | Frontiers in Cardiovascular Medicine |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fcvm.2022.903660/full |
| _version_ | 1817999049636184064 |
|---|---|
| author | Xixiang Lin Xixiang Lin Feifei Yang Yixin Chen Xiaotian Chen Wenjun Wang Xu Chen Xu Chen Qiushuang Wang Liwei Zhang Huayuan Guo Bohan Liu Liheng Yu Haitao Pu Peifang Zhang Zhenzhou Wu Xin Li Daniel Burkhoff Kunlun He |
| author_facet | Xixiang Lin Xixiang Lin Feifei Yang Yixin Chen Xiaotian Chen Wenjun Wang Xu Chen Xu Chen Qiushuang Wang Liwei Zhang Huayuan Guo Bohan Liu Liheng Yu Haitao Pu Peifang Zhang Zhenzhou Wu Xin Li Daniel Burkhoff Kunlun He |
| author_sort | Xixiang Lin |
| collection | DOAJ |
| description | ObjectiveTo compare the performance of a newly developed deep learning (DL) framework for automatic detection of regional wall motion abnormalities (RWMAs) for patients presenting with the suspicion of myocardial infarction from echocardiograms obtained with portable bedside equipment versus standard equipment.BackgroundBedside echocardiography is increasingly used by emergency department setting for rapid triage of patients presenting with chest pain. However, compared to images obtained with standard equipment, lower image quality from bedside equipment can lead to improper diagnosis. To overcome these limitations, we developed an automatic workflow to process echocardiograms, including view selection, segmentation, detection of RWMAs and quantification of cardiac function that was trained and validated on image obtained from bedside and standard equipment.MethodsWe collected 4,142 examinations from one hospital as training and internal testing dataset and 2,811 examinations from other hospital as the external test dataset. For data pre-processing, we adopted DL model to automatically recognize three apical views and segment the left ventricle. Detection of RWMAs was achieved with 3D convolutional neural networks (CNN). Finally, DL model automatically measured the size of cardiac chambers and left ventricular ejection fraction.ResultsThe view selection model identified the three apical views with an average accuracy of 96%. The segmentation model provided good agreement with manual segmentation, achieving an average Dice of 0.89. In the internal test dataset, the model detected RWMAs with AUC of 0.91 and 0.88 respectively for standard and bedside ultrasound. In the external test dataset, the AUC were 0.90 and 0.85. The automatic cardiac function measurements agreed with echocardiographic report values (e. g., mean bias is 4% for left ventricular ejection fraction).ConclusionWe present a fully automated echocardiography pipeline applicable to both standard and bedside ultrasound with various functions, including view selection, quality control, segmentation, detection of the region of wall motion abnormalities and quantification of cardiac function. |
| first_indexed | 2024-04-14T03:02:45Z |
| format | Article |
| id | doaj.art-796301610bcc4c08b7e03cc1e108c47f |
| institution | Directory Open Access Journal |
| issn | 2297-055X |
| language | English |
| last_indexed | 2024-04-14T03:02:45Z |
| publishDate | 2022-08-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Cardiovascular Medicine |
| spelling | doaj.art-796301610bcc4c08b7e03cc1e108c47f2022-12-22T02:15:51ZengFrontiers Media S.A.Frontiers in Cardiovascular Medicine2297-055X2022-08-01910.3389/fcvm.2022.903660903660Echocardiography-based AI detection of regional wall motion abnormalities and quantification of cardiac function in myocardial infarctionXixiang Lin0Xixiang Lin1Feifei Yang2Yixin Chen3Xiaotian Chen4Wenjun Wang5Xu Chen6Xu Chen7Qiushuang Wang8Liwei Zhang9Huayuan Guo10Bohan Liu11Liheng Yu12Haitao Pu13Peifang Zhang14Zhenzhou Wu15Xin Li16Daniel Burkhoff17Kunlun He18Medical Big Data Center, Chinese PLA General Hospital, Beijing, ChinaMedical School of Chinese PLA, Beijing, ChinaMedical Big Data Center, Chinese PLA General Hospital, Beijing, ChinaBioMind Technology, Beijing, ChinaBioMind Technology, Beijing, ChinaMedical Big Data Center, Chinese PLA General Hospital, Beijing, ChinaMedical Big Data Center, Chinese PLA General Hospital, Beijing, ChinaMedical School of Chinese PLA, Beijing, ChinaFourth Medical Center of PLA General Hospital, Beijing, ChinaFourth Medical Center of PLA General Hospital, Beijing, ChinaMedical Big Data Center, Chinese PLA General Hospital, Beijing, ChinaMedical Big Data Center, Chinese PLA General Hospital, Beijing, ChinaMedical Big Data Center, Chinese PLA General Hospital, Beijing, ChinaBioMind Technology, Beijing, ChinaBioMind Technology, Beijing, ChinaBioMind Technology, Beijing, ChinaSixth Medical Center of PLA General Hospital, Beijing, ChinaCardiovascular Research Foundation, New York, NY, United StatesMedical Big Data Center, Chinese PLA General Hospital, Beijing, ChinaObjectiveTo compare the performance of a newly developed deep learning (DL) framework for automatic detection of regional wall motion abnormalities (RWMAs) for patients presenting with the suspicion of myocardial infarction from echocardiograms obtained with portable bedside equipment versus standard equipment.BackgroundBedside echocardiography is increasingly used by emergency department setting for rapid triage of patients presenting with chest pain. However, compared to images obtained with standard equipment, lower image quality from bedside equipment can lead to improper diagnosis. To overcome these limitations, we developed an automatic workflow to process echocardiograms, including view selection, segmentation, detection of RWMAs and quantification of cardiac function that was trained and validated on image obtained from bedside and standard equipment.MethodsWe collected 4,142 examinations from one hospital as training and internal testing dataset and 2,811 examinations from other hospital as the external test dataset. For data pre-processing, we adopted DL model to automatically recognize three apical views and segment the left ventricle. Detection of RWMAs was achieved with 3D convolutional neural networks (CNN). Finally, DL model automatically measured the size of cardiac chambers and left ventricular ejection fraction.ResultsThe view selection model identified the three apical views with an average accuracy of 96%. The segmentation model provided good agreement with manual segmentation, achieving an average Dice of 0.89. In the internal test dataset, the model detected RWMAs with AUC of 0.91 and 0.88 respectively for standard and bedside ultrasound. In the external test dataset, the AUC were 0.90 and 0.85. The automatic cardiac function measurements agreed with echocardiographic report values (e. g., mean bias is 4% for left ventricular ejection fraction).ConclusionWe present a fully automated echocardiography pipeline applicable to both standard and bedside ultrasound with various functions, including view selection, quality control, segmentation, detection of the region of wall motion abnormalities and quantification of cardiac function.https://www.frontiersin.org/articles/10.3389/fcvm.2022.903660/fullartificial intelligence - AImyocardial infarctionechocardiographydeep learningbedside ultrasound |
| spellingShingle | Xixiang Lin Xixiang Lin Feifei Yang Yixin Chen Xiaotian Chen Wenjun Wang Xu Chen Xu Chen Qiushuang Wang Liwei Zhang Huayuan Guo Bohan Liu Liheng Yu Haitao Pu Peifang Zhang Zhenzhou Wu Xin Li Daniel Burkhoff Kunlun He Echocardiography-based AI detection of regional wall motion abnormalities and quantification of cardiac function in myocardial infarction Frontiers in Cardiovascular Medicine artificial intelligence - AI myocardial infarction echocardiography deep learning bedside ultrasound |
| title | Echocardiography-based AI detection of regional wall motion abnormalities and quantification of cardiac function in myocardial infarction |
| title_full | Echocardiography-based AI detection of regional wall motion abnormalities and quantification of cardiac function in myocardial infarction |
| title_fullStr | Echocardiography-based AI detection of regional wall motion abnormalities and quantification of cardiac function in myocardial infarction |
| title_full_unstemmed | Echocardiography-based AI detection of regional wall motion abnormalities and quantification of cardiac function in myocardial infarction |
| title_short | Echocardiography-based AI detection of regional wall motion abnormalities and quantification of cardiac function in myocardial infarction |
| title_sort | echocardiography based ai detection of regional wall motion abnormalities and quantification of cardiac function in myocardial infarction |
| topic | artificial intelligence - AI myocardial infarction echocardiography deep learning bedside ultrasound |
| url | https://www.frontiersin.org/articles/10.3389/fcvm.2022.903660/full |
| work_keys_str_mv | AT xixianglin echocardiographybasedaidetectionofregionalwallmotionabnormalitiesandquantificationofcardiacfunctioninmyocardialinfarction AT xixianglin echocardiographybasedaidetectionofregionalwallmotionabnormalitiesandquantificationofcardiacfunctioninmyocardialinfarction AT feifeiyang echocardiographybasedaidetectionofregionalwallmotionabnormalitiesandquantificationofcardiacfunctioninmyocardialinfarction AT yixinchen echocardiographybasedaidetectionofregionalwallmotionabnormalitiesandquantificationofcardiacfunctioninmyocardialinfarction AT xiaotianchen echocardiographybasedaidetectionofregionalwallmotionabnormalitiesandquantificationofcardiacfunctioninmyocardialinfarction AT wenjunwang echocardiographybasedaidetectionofregionalwallmotionabnormalitiesandquantificationofcardiacfunctioninmyocardialinfarction AT xuchen echocardiographybasedaidetectionofregionalwallmotionabnormalitiesandquantificationofcardiacfunctioninmyocardialinfarction AT xuchen echocardiographybasedaidetectionofregionalwallmotionabnormalitiesandquantificationofcardiacfunctioninmyocardialinfarction AT qiushuangwang echocardiographybasedaidetectionofregionalwallmotionabnormalitiesandquantificationofcardiacfunctioninmyocardialinfarction AT liweizhang echocardiographybasedaidetectionofregionalwallmotionabnormalitiesandquantificationofcardiacfunctioninmyocardialinfarction AT huayuanguo echocardiographybasedaidetectionofregionalwallmotionabnormalitiesandquantificationofcardiacfunctioninmyocardialinfarction AT bohanliu echocardiographybasedaidetectionofregionalwallmotionabnormalitiesandquantificationofcardiacfunctioninmyocardialinfarction AT lihengyu echocardiographybasedaidetectionofregionalwallmotionabnormalitiesandquantificationofcardiacfunctioninmyocardialinfarction AT haitaopu echocardiographybasedaidetectionofregionalwallmotionabnormalitiesandquantificationofcardiacfunctioninmyocardialinfarction AT peifangzhang echocardiographybasedaidetectionofregionalwallmotionabnormalitiesandquantificationofcardiacfunctioninmyocardialinfarction AT zhenzhouwu echocardiographybasedaidetectionofregionalwallmotionabnormalitiesandquantificationofcardiacfunctioninmyocardialinfarction AT xinli echocardiographybasedaidetectionofregionalwallmotionabnormalitiesandquantificationofcardiacfunctioninmyocardialinfarction AT danielburkhoff echocardiographybasedaidetectionofregionalwallmotionabnormalitiesandquantificationofcardiacfunctioninmyocardialinfarction AT kunlunhe echocardiographybasedaidetectionofregionalwallmotionabnormalitiesandquantificationofcardiacfunctioninmyocardialinfarction |